The problem of automatic semantic features mapping is an important aspect of natural language processing with a wide variety of applications. In the paper we compare the type, amount and structure of information available in two Polish wordnets, namely plWordNet and Polnet. In particular we analyze the ability of both semantic data resources to become a reliable source of expert knowledge aiding the process of semantic features mapping. The comparison is based on selected metrics obtained for the binary word-feature classification problem as well as some statistical measures describing wordnets’ contents. The experimental results are presented and explained in detail along with an outline of further research perspectives.